Can you see something hidden in the images below? Squint your eye until you see it! Stable Diffusion AI is particularly suited for generating seemingly innocent images with hidden words.
The best tools to generate this kind of image are ControlNet and ADetailer. The image itself is controlled by the text prompt. It is infinitely customizable and always unique.
In this article, you will find a step-by-step guide with detailed settings to generating these images in AUTOMATIC1111.
Hi Andrew,
thanks for the interesting workflow.
I tried to run this workflow in ComfyUI, but no matter how I adjusted the Control Weight or the Starting/Ending Control Step, I couldn’t get an image that blended the text and the image well.
I also tried to set up and run the same in SD-Forge, but I didn’t get a good image either.
Can you give me some advice?
I just tried on forge and the setting is very sensitive. I can kind of get something close, but not as good as the ones I got with A1111.
Setting “my prompt is more important” and “pixel perfect” helps to reduce the sensitivity.
Look on all images in post, don’t recognize any text, lol.
That’s the definition of “hidden” XD
Sorry for the joke, every image has a hidden “WOW!” text.
Squint your eyes harder lol!